Journal of Experimental Botany, Vol. 70, No. 16 pp.

4139–4154, 2019
doi:10.1093/jxb/erz213  Advance Access Publication 4 May, 2019

REVIEW PAPER

Rhodanese domain-containing sulfurtransferases:

multifaceted proteins involved in sulfur trafficking in plants
Benjamin Selles, Anna Moseler, Nicolas Rouhier and Jérémy Couturier*,
Université de Lorraine, Inra, IAM, F-54000 Nancy, France

Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023

*  Correspondence: jeremy.couturier@univ-lorraine.fr

Received 17 April 2019; Editorial decision 29 April 2019; Accepted 29 April 2019

Editor: Stanislav Kopriva, University of Cologne, Germany

Abstract
Sulfur is an essential element for the growth and development of plants, which synthesize cysteine and methionine
from the reductive assimilation of sulfate. Besides its incorporation into proteins, cysteine is the building block for the
biosynthesis of numerous sulfur-containing molecules and cofactors. The required sulfur atoms are extracted either
directly from cysteine by cysteine desulfurases or indirectly after its catabolic transformation to 3-mercaptopyruvate,
a substrate for sulfurtransferases (STRs). Both enzymes are transiently persulfidated in their reaction cycle, i.e. the
abstracted sulfur atom is bound to a reactive cysteine residue in the form of a persulfide group. Trans-persulfidation
reactions occur when sulfur atoms are transferred to nucleophilic acceptors such as glutathione, proteins, or small
metabolites. STRs form a ubiquitous, multigenic protein family. They are characterized by the presence of at least
one rhodanese homology domain (Rhd), which usually contains the catalytic, persulfidated cysteine. In this review,
we focus on Arabidopsis STRs, presenting the sequence characteristics of all family members as well as their bio-
chemical and structural features. The physiological functions of particular STRs in the biosynthesis of molybdenum
cofactor, thio-modification of cytosolic tRNAs, arsenate tolerance, cysteine catabolism, and hydrogen sulfide forma-
tion are also discussed.

Keywords:   Cysteine, hydrogen sulfide signaling, persulfide group, rhodanese, sulfur trafficking, sulfurtransferase.

Introduction
Unlike animals, which cannot assimilate inorganic sulfur and
thus need to absorb cysteine or methionine, plants achieve the
reductive assimilation of sulfate to synthesize cysteine, forming
sulfite and sulfide as plastidial intermediates, and subsequently
methionine and glutathione (GSH) (Takahashi et  al., 2011).
Other key cellular components, such as vitamins (biotin, thia-
mine, lipoic acid), tRNAs containing thionucleosides, or co-
factors such as iron–sulfur (Fe–S) clusters and molybdenum
cofactor (Moco), contain sulfur atoms coming from various
origins. In the case of vitamin or camalexin biosynthesis, sulfur
is respectively provided by the degradation of Fe–S clus-
ters (Lanz and Booker, 2015) or by a GSH molecule via the
action of glutathione transferase (Su et al., 2011). In the case of
thionucleoside formation or Fe–S cluster and Moco synthesis,
other sulfur sources are used (Mueller, 2006). For instance, the
pyridoxal 5′-phosphate-dependent cysteine desulfurases (CDs)
catalyse the desulfuration of cysteine (Zheng et  al., 1993),
leading to the formation of a persulfide enzyme intermediate
and the concomitant release of alanine (Behshad and Bollinger,
2009). The efficiency and specificity of sulfur transfer to ac-
ceptor molecules vary according to the subclass of CDs and
the type of sulfur acceptors, the nature of which dictates the
direction and flow of the sulfur transfer reaction. Whereas
glutathione persulfide (GSSH) may represent a relatively stable

© The Author(s) 2019. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved.
For permissions, please email: journals.permissions@oup.com
4140 | Selles et al.
and specific transport form of sulfur atoms (Hildebrandt and
Grieshaber, 2008; Ida et al., 2014), carrier proteins ensure the
specific transfer of sulfur to acceptors. This function is pri-
marily achieved by sulfurtransferases (STRs; E.C.2.8.1.x).
They possess one or several rhodanese (Rhd) domains usu-
ally including a reactive catalytic cysteine (Hatzfeld and Saito,
2000; Bordo and Bork, 2002; Cipollone et al., 2007; Guretzki
and Papenbrock, 2011). Hence, most STRs catalyse the transfer
of a sulfur atom from donors to nucleophilic acceptors by
forming intermediate protein persulfides on a reactive cata-
lytic cysteine.When two polypeptides are involved, the transfer
of persulfide groups is referred to as a trans-persulfidation
reaction. Although STRs are ubiquitously distributed, they
differ quite considerably in terms of primary sequences, pro-
tein domain architecture, and structure/length of the active
site loop (Bordo and Bork, 2002). Nonetheless, three major
groups can be differentiated (Bordo and Bork, 2002; Cipollone
et  al., 2007; Bartels et  al., 2007). The first one comprises
STRs formed by a single Rhd domain that include so-called
CDC25 phosphatases and thiosulfate sulfurtransferases (TSTs;
EC.2.8.1.1) according to their ability to use thiosulfate as a
sulfur donor in vitro. The presence of an additional residue in
the active-site loop of CDC25 phosphatases (seven residues
compared with six in TSTs) constitutes an important differ-
ence involved in the capacity of CDC25 to dephosphorylate
tyrosine residues (Bordo and Bork, 2002). The second major
group is formed by STRs containing two Rhd domains
but only the C-terminal one possesses the catalytic cyst-
eine. They are present in most organisms. According to their
ability to use 3-mercaptopyruvate (3-MP) as a substrate in
vitro, they are referred to as 3-MP-STRs or mercaptopyruvate
sulfurtransferases (MSTs; EC.2.8.1.2) (Nakamura et al., 2000;
Papenbrock and Schmidt, 2000; Colnaghi et  al., 2001; Yadav
et  al., 2013). It is worth mentioning that bacteria and mam-
mals possess additional STR isoforms with two Rhod domains
(Bordo and Bork, 2002; Nambi et al., 2015), such as the bo-
vine rhodanese protein named Rhobov, which display TST
activity (Blumenthal and Heinrikson, 1971). The mammalian
rhodanese isoforms have a CRXGX[R/T] active site signa-
ture instead of the characteristic CG[S/T]GVT motif found in
MSTs (Bordo and Bork, 2002; Cipollone et al., 2007). In fact,
interchanging the two residues following the catalytic cysteine
of rat rhodanese and MST by site-directed mutagenesis was
sufficient to change the activity profile from one to the other
(Nagahara et  al., 1995; Nagahara and Nishino, 1996). Finally,
the third main group is formed by STRs containing a Rhd do-
main possessing the catalytic cysteine fused to a domain having
another function conferring them specific roles (Bordo and
Bork, 2002; Cipollone et al., 2007).
The recent comparative analysis of 25 genomes of photosyn-
thetic organisms has provided a detailed classification and in-
formation about evolutionary features of STRs pointing to the
expansion of this family in higher plants. Hence, in photosyn-
thetic organisms, the STR family comprises 5–35 genes, and
was divided into nine clusters according to the STR primary
sequences and domain arrangements (Moseler et  al., 2019).
Plant STR isoforms of clusters I  (MSTs), VII (multidomain
STRs), VIII (TSTs) and IX (TSTs) possess orthologs in most
organisms whereas other clusters are specific (clusters III and
V) or contain an increased number of representatives in plants
(clusters II, IV and VI). In the context of this review on the in-
volvement of plant STRs in sulfur transfer reactions, those that
belong to cluster II and do not possess the conserved catalytic
cysteine in the Rhd domain will not be described in detail.
Hence, we discuss the recent genetic, biochemical, structural,
and molecular studies that have contributed to improving our
understanding of the physiological roles of STRs in plants.
After illustrating the diversity of plant STRs in terms of pri-
mary sequences, protein domain architecture, subcellular lo-
calization and gene expression using Arabidopsis isoforms as
representatives, we focus our attention on the documented
roles of STRs in molybdopterin synthesis, thionucleoside
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
modification, arsenate tolerance, cysteine catabolism, and the
biosynthesis of hydrogen sulfide (H2S), a signaling molecule
promoting protein persulfidation.
Diversity of plant STR isoforms: focus on
Arabidopsis
The Arabidopsis nuclear genome was initially reported to en-
code 20 STR sequences (Papenbrock et al., 2011). However, a
recent careful analysis led us to include an additional protein
containing a Rhd domain, which was numbered consecutively
as AtSTR19 (At3g59780), even though it does not have the
conserved cysteine residue (Moseler et al., 2019).
Domain organization, expression, and subcellular
localization
According to the variability observed among STRs in terms
of primary sequence and domain organization (including po-
tential transmembrane spans), the Arabidopsis STR equipment
forms nine different clusters (Table 1; Moseler et  al., 2019).
STRs composed of two Rhd domains form the first group
comprising so-called MSTs, i.e. STR1 and STR2, which
are 379 and 342 residues long, respectively (Table 1). STR3
(387 residues), STR4 (466 residues), STR4a (264 residues),
and STR19 (686 residues) form cluster II as they contain a
single Rhd domain associated with one, two, one, and three
predicted transmembrane regions, respectively (Table 1). Note
that STR3, STR4, and STR19 are considered as inactive
STRs because the catalytic cysteine is replaced by an aspar-
tate. Cluster III contains the sole STR5 isoform. It possesses
a single Rhd domain exhibiting the catalytic cysteine but in a
quite divergent HCALSQVR signature typical of the general
HCX5R signature of CDC25 phosphatases. STR6, STR7, and
STR8 belong to cluster IV. They have also a single Rhd do-
main with a strictly conserved CTGGIR signature, but they
are elongated (581, 474, and 448 residues respectively) on both
sides of the Rhd domain.The next group (cluster V) comprises
STR9, STR10, and STR11 isoforms (214–292 residues) cor-
responding to proteins with a single Rhd domain exhibiting a
catalytic cysteine and a predicted C-terminal transmembrane
segment (Table 1). STR12 (299 residues) and STR13 (464 res-
idues) form two additional groups (clusters VI and VII), having
 Roles of sulfurtransferases in sulfur trafficking in plants  |  4141

Table 1.  Subcellular localization and domain organization of the STR/rhodanese family members from Arabidopsis

Name (ac- Cluster Length Active Domain arrangement Subcellular localization

cession no.) (aa) site motif
STR1 I 379 CGTGVT Mitochondrion
(At1g79230) GFP fusion (Hatzfeld and Saito, 2000; Nakamura et al.,
2000; Bauer et al., 2004) and immunoblot analysis
(Nakamura et al., 2000; Bauer et al., 2004)
STR2 I 342 CGTGVT Cytosol
(At1g16460) GFP fusion (Hatzfeld and Saito, 2000; Nakamura et al.,
2000; Bauer et al., 2004) and immunoblot analysis
(Nakamura et al., 2000; Bauer et al., 2004)
STR3 II 387 DSYTDS Chloroplast (thylakoid membrane)
(At5g23060) GFP/YFP fusion (Weinl et al., 2008; Nomura et al.,
2008; Vainonen et al., 2008), immunoblot analysis

Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023

(Weinl et al., 2008; Vainonen et al., 2008) and prote-
omics (Peltier et al., 2004)
STR4/TROL II 466 DKFDGN Chloroplast (inner membrane envelope and thyla-
(At4g01050) koid)
YFP fusion (Jurić et al., 2009) and proteomics (Friso
et al., 2004; Peltier et al., 2004)
STR4a II 264 CVLDNF Chloroplast (thylakoid)
(At3g25480) Proteomics (Peltier et al., 2004; Tomizioli et al., 2014)
STR5/ III 146 CALSQVR Mitochondrion
CDC25/ACR2 GFP fusion (Narsai et al., 2011)
(At5g03455)
STR6 IV 581 CTGGIR Unknown
(At1g09280)
STR7 IV 474 CTGGIR Unknown
(At2g40760)
STR8 IV 448 CTGGIR Chloroplast (thylakoid membrane)
(At1g17850) Proteomics (Tomizioli et al., 2014)
STR9 V 234 CQEGLR Chloroplast (thylakoid membrane)
(At2g42220) Proteomics (Tomizioli et al., 2014)
STR10 V 214 CGEGLR Mitochondrion
(At3g08920) GFP fusion (Bartels, 2006)
STR11 V 292 CQKGLR Chloroplast (thylakoid membrane)
(At4g24750) GFP fusion (Bartels, 2006) and proteomics (Tomizioli
et al., 2014)
STR12 VI 299 CKVGGR Chloroplast (stroma)
(At5g19370) Proteomics (Tomizioli et al., 2014)
STR13/CNX5 VII 464 CRRGND Cytosol
(At5g55130) Bimolecular fluorescence complementation (Kaufholdt
et al., 2013)
STR14 VIII 224 CSSAGT Chloroplast (envelope, thylakoid)
(At4g27700) GFP fusion (Bauer et al., 2004)
STR15/Sen1 IX 182 CESGQM Chloroplast (thylakoid membrane)
(At4g35770) GFP fusion (Bauer et al., 2004) and electronic micros-
copy (Bauer et al., 2004)
STR16 IX 120 CQSGGR Chloroplast
(At5g66040) GFP fusion (Bauer et al., 2004)
STR17 IX 156 CKSGVR Unknown
(At2g17850)
STR17a/ IX 169 CNAGGR Cytosol/nucleus
ARQ1/HAC1 GFP fusion (Sánchez-Bermejo et al., 2014)
(At2g21045)
STR18 IX 136 CQSGAR Cytosol
(At5g66170) GFP fusion (Bauer et al., 2004)
STR19 II 686 DADGTR Chloroplast
(At3g59780) Proteomics (Peltier et al., 2004; Tomizioli et al., 2014)

The catalytic cysteine is indicated in bold. Domain prediction is based on Pfam (http://pfam.xfam.org/) and Interpro (http://www.ebi.ac.uk/interpro/).
FSH1, serine hydrolase; MoeB, molybdopterin biosynthesis; Rota, rotamase.
4142 | Selles et al.
an N-terminal rotamase and MoeB domain, respectively, in
addition to the active Rhd domain (Table 1). Cluster VIII con-
tains only AtSTR14 because the Rhd domain has a specific
CSSAGT signature and it displays a short N-terminal extension
forming a protein of 224 residues. Finally, cluster IX comprises
STR15–STR18. They are all short STR versions of less than
182 amino acids, the difference residing in the active site sig-
nature and in the presence of N-terminal targeting sequences,
but none has additional recognizable domains, except a pre-
dicted C-terminal transmembrane segment in STR15 likely
promoting its localization to thylakoids (Bauer et al., 2004).
As an additional criterion to understand the diversity and evo-
lution of the STR family in plants, the gene expression pattern
and protein subcellular localization of Arabidopsis STRs was
analysed. A recent, high resolution developmental transcript pro-
file was analysed (Winter et al., 2007; Klepikova et al., 2016) and
selected tissues and stages were represented as a heat map (Fig.
1). Three main clusters are distinguished. A first cluster regroups
STR2, STR5, STR15, and STR16 representatives. They are
mostly expressed in leaves and hypocotyl/stems (Fig. 1). STRs
present in the second cluster (STR3, STR4, STR4a, STR7, STR8,
STR9, STR10, STR11, STR12, and STR14) have in common a
higher expression level in cotyledons, mature leaves (8–12) and
vegetative rosette before or after the transition to flowering (Fig.
1). Two sub-clusters can be distinguished mainly based on their
differential expression in seeds. Consistently, all STRs that have
Fig. 1.  Heatmap representation of STR transcript profiles at selected developmental stages and tissues of Arabidopsis. Gene expression data for each
Arabidopsis STR were retrieved from the ‘Developmental Map’ tab of the Arabidopsis eFP Browser database (http://bar.utoronto.ca/efp/cgi-bin/efpWeb.
cgi) (Winter et al., 2007; Klepikova et al., 2016). No data were available for STR19. Numerical data were extracted and represented as a heatmap
using the expression tool of Heatmapper software (http://www2.heatmapper.ca/expression/) (Babicki et al., 2016). The parameters applied are score
representation by row, clustering method by average linkage, and Pearson distance measurement.
been demonstrated or predicted as targeted to the chloroplasts/
plastids (Bauer et al., 2004; Friso et al., 2004; Peltier et al., 2004;
Bartels, 2006; Nomura et al., 2008; Vainonen et al., 2008; Weinl
et al., 2008; Jurić et al., 2009; Tomizioli et al., 2014) are encoded
by genes present in these two clusters (Table 1; Fig. 1). With re-
gards to genes present in the third cluster (STR1, STR6, STR13,
STR17, STR17a, and STR18), their expression does not vary
much among organs but they have in common a highest tran-
script level in dry and imbibed seeds and in roots. Note that
AtSTR17a is specifically more expressed in roots. The proteins
encoded by these genes are expressed in the cytosol (STR13,
STR17a, and STR18) (Bauer et  al., 2004), in mitochondria
(STR1) (Hatzfeld and Saito, 2000; Nakamura et al., 2000; Bauer
et al., 2004), or have unknown localizations (STR6 and STR17)
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
having no typical targeting sequences. These diverse subcellular
localizations and gene expression patterns are understandable
considering the existence of numerous sulfur-containing com-
pounds in the different subcellular compartments of plant cells
and thus the need of specific systems to mobilize and transfer
sulfur for their biosynthesis or modification.
Biochemical and structural properties
Despite the described diversity of STR primary
sequences, the fold adopted by the Rhd domain consists
of a β1α1β2α2β3α3β4α4β5α5 topology organizing into a
STR16
STR5
STR2
STR15
STR11
STR4a
STR12
STR7
STR8
STR14
STR4
STR9
STR10
STR3
STR18
STR17
STR13
STR1
STR6
STR17a
 Roles of sulfurtransferases in sulfur trafficking in plants  |  4143

well-conserved central β-sheet composed of five strands sur-
rounded by four to five α-helices (Ploegman et al., 1978). Up
to now, three-dimensional structures have been solved for
AtSTR5/CDC25, AtSTR16, and the inactive Rhd domain of
AtSTR4/TROL (Landrieu et al., 2004a; Pantoja-Uceda et al.,
2005; Cornilescu et al., 2006). A structure-based amino acid
alignment of Arabidopsis STRs possessing the catalytic cyst-
eine, using AtSTR16 NMR structure (PDB 1TQ1) as a tem-
plate, indicates that only a few residues are strictly conserved
(Fig. 2A). In the N-terminal part of the Rhd domain, a three
residue motif, D28[V/A]29R30 (AtSTR16 numbering unless
otherwise indicated), is highly conserved in Arabidopsis
paralogs (Fig. 2A). It is also conserved in representatives of
the STR5/CDC25 family and more generally in the Rhd
domains of all plant STRs (Moseler et al., 2019). The residues
are located at the end of β2 (Fig. 2B) and were proposed
to participate in AtSTR16 stability (Cornilescu et al., 2006).
The catalytic cysteine (Cys80) is strictly conserved and is pre-
sent in the loop between β4 and α4 secondary structures and
forms with the five or six (in the case of CDC25) following
amino acids the so-called active site loop of Rhd domain (Figs
2B, 3). The residues forming the active site are quite variable.

Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023

Fig. 2.  Structure-based sequence alignment of Rhd domains of Arabidopsis STR isoforms. The sequences corresponding to the Rhd domain of
Arabidopsis STRs possessing a cysteine residue were aligned with MUSCLE (https://www.ebi.ac.uk/Tools/msa/muscle/). Colored alignment (A) and the
corresponding structure representing the structural conservation level at each amino acid position (B) were generated by CONSURF (http://consurf.tau.
ac.il/2016/) using the PDB file of the AtSTR16 NMR structure (PDB 1TQ1). The sequence numbering and secondary structures correspond to AtSTR16.
In (A) the catalytic cysteine residue is marked by an asterisk and amino acids studied by mutagenesis indicated by triangles. In (B) the residues studied
by mutagenesis in various selected STR representatives are shown with their lateral chains as stick and labeled. The most conserved structural positions
(D28[V/A]29R30, C80, G82, L93, G97, and G107) are also labeled whereas green parts correspond to regions with insufficient data to represent a conservation.
4144 | Selles et al.
Val251
Gly250
Rhobov / TST Thr252
Lys249
CRKGVT
Cys247
Arg248
Arg85
AtSTR16 / TST
Gly83
Ser82 Gly84
CQSGGR
Gln81 Cys80
Fig. 3.  Structural comparison of the active site loops of TST-type and MST-type STRs. The representation of the six amino acids forming the so-called
active site loop from bovine Rhobov (PDB 1RHD), human 3-MST (PDB 4JGT), Arabidopsis STR16 (PDB 1TQ1), and human TSTD (PDB 6BEV) was
generated using PyMOL.
A characteristic signature was defined for MSTs (CGSGVT)
and TSTs (CRKGVT) and it is likely that this relates at least
in part to electrostatic differences of their substrates, as ex-
emplified by 3-MP and thiosulfate, respectively (Bordo and
Bork, 2002). At positions +1 and +2 after the catalytic cyst-
eine, bulky and charged residues (Arg/Lys) are found in TSTs
whereas smaller and uncharged residues (Gly/Ser) are pre-
sent in MSTs (Fig. 3). The Gly residue usually at position +3
of the catalytic cysteine and the Leu residue at position 93
(Fig. 2A) are conserved in all Arabidopsis STRs even though
the AtSTR5 signature containing the catalytic cysteine does
not align at all. Two other Gly residues at positions 97 and
107 are present in all plant STR isoforms except in STR1/2
orthologs in which Gly107 is replaced by a serine. These three
Gly residues are located just before and at the end of α4,
and at the initiation of α5, respectively (Fig. 2B). Despite the
high conservation of these four residues (Gly83, Leu93, Gly97,
Gly107), their roles in STR structure/function are unknown.
On the contrary, the importance of other residues was inves-
tigated more deeply using both plant and non-plant STRs
(Table 2). Hence, besides the mandatory catalytic cysteine
(Cys80 in Fig. 2), other residues are important for the catalytic
activity of the Rhd domain. For example, the change of the
cysteine residue close to the active site in AtSTR1 (Cys339,
position 87) to a valine residue led to a 4-fold decrease of the
TST activity whereas the MST activity was slightly reduced
(Burow et al., 2002). However, although highly conserved in
MSTs from terrestrial plants, this cysteine is not present in
other STRs. For instance, it corresponds to Ile87 of AtSTR16,
a position presenting a weak conservation score (Fig. 2A).
Substitutions of the two other Cys residues, Cys294 to Asn
and Cys304 to Asp, found in the active C-terminal Rhd do-
main of AtSTR1 had minor effects with a slight decrease
of both Km and kcat towards thiosulfate (Burow et al., 2002).
According to the alignment, the corresponding amino acids
in AtSTR16 (positions 43 and 53) are found respectively in
the α2–β3 and β3–α3 loops and exhibit a low conservation
score (Fig. 2B).
Val252
Gly251
Thr253 Human 3-MST / MST
Ser250 CGSGVT
Cys248
Gly249
Lys83
Gly82 Human TSTD / TST
Met81
CQMGKR
Arg84
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
Cys79
Gln80
As the biochemical data concerning plant STRs are rela-
tively scarce, we consider here mutagenesis studies conducted
on non-plant STR orthologs such as bovine rhodanese
(Rhobov), the Escherichia coli MST ortholog SseA, the
Azotobacter vinelandii RhdA protein and the Rhd domain of
human MOCS3 to identify other residues potentially im-
portant for the structure and activity of plant STRs (Table 2).
Structural analysis of human MST and biochemical charac-
terization of E. coli SseA indicated that two arginine residues,
Arg178 and Arg187 (position 34 and insertion between positions
36 and 37, respectively), play a role in substrate recognition,
being involved in the proper positioning of the substrate rela-
tive to the catalytic Cys237 (Figs 2A, 3) (Yadav et al., 2013; Lec
et  al., 2018). Hence in EcSseA, the change of Arg178 to Leu
leads to a decrease of protein affinity towards 3-MP (Lec et al.,
2018). Both AtSTR1 and AtSTR2 (and AtSTR5) possess this
Arg residue whereas most other STRs possess a Glu residue
(Fig. 2A). On the contrary, Arg187 is only present in AtSTR1
and is located into an insertion sequence specific to MST
orthologs (Fig. 2). Interestingly, the mutation of Arg187, which
is part of the hydrophobic pocket of the active site (Ploegman
et al., 1978), into a Leu in bovine Rhobov, a TST-type STR,
leads to a 20-fold decrease of affinity towards thiosulfate (Km
shifted from 3.7 to 73.2 mM in the R187L variant) (Luo and
Horowitz, 1994). Hence, this residue may also be important for
thiosulfate accommodation in the active site.
It is expected that other residues found in the active site
loops of MST, TST, or CDC25 generate the necessary se-
lectivity of these proteins towards their respective substrates
(Bordo and Bork, 2002; Cipollone et al., 2007; Lec et al., 2018;
Libiad et al., 2018).Thus, apart from the Arg residues described
above, the importance of the residues present at positions +2
and +5 of the catalytic Cys in regular STRs was investigated.
For MST-type STR proteins, studies performed using
EcSseA, human and Leishmania major MST variants initially
suggested that the +2 residue (Ser82 in Figs 2A, 3) is involved in
the deprotonation of 3-MP and in the stabilization of the pyru-
vate enolate (Colnaghi et al., 2001; Alphey et al., 2003; Huang
 Roles of sulfurtransferases in sulfur trafficking in plants  |  4145

Table 2.  Mutagenesis studies performed using STR isoforms

Protein Uniprot Ref Type Position in Corresponding Substi- Catalytic effect Reference
name the original position in tution
sequence AtSTR16
Rhobov P00586 TST 187 34 R→L Decreased reactivity to- Luo and Horowitz
wards thiosulfate (1994)
SseA P31142 MST 178 34 R→L Decreased reactivity to- Lec et al. (2018)
wards 3-MP
STR1 O64530 MST 294 43 C→N Slightly reduced TST ac- Burow et al. (2002)
tivity, no impact on MST
activity
STR1 O64530 MST 304 53 C→E Slightly reduced TST ac- Burow et al. (2002)
tivity, no impact on MST
activity

Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023

STR1 O64530 MST 333 80 C→S Abolished TST and MST Burow et al. (2002)
activities
MOCS3 O95396 — 412 80 C→A Abolished STR activity Matthies et al.
(2004)
RhdA P52197 TST 232 82 T→K Increased TST activity Pagani et al. (2000)
RhdA P52198 TST 232 82 T→A Increased TST activity Pagani et al. (2000)
Rhobov P00586 TST 250 82 K→A Abolished TST activity Luo and Horowitz
(1994)
SseA P31142 MST 239 82 S→A Decreased affinity for Colnaghi et al.
3-MP; abolished thiosul- (2001), Lec et al.
fate binding (2018)
SseA P31142 MST 239 82 S→K Decreased MST activity; Colnaghi et al.
increased affinity for thio- (2001)
sulfate
MOCS3 O95396 — 417 85 D→R 470-fold increased activity Krepinsky and
Leimkühler (2007)
MOCS3 O95396 — 417 85 D→T 90-fold increased activity Krepinsky and
Leimkühler (2007)
STR1 O64530 MST 339 87 C→V 4-fold decreased TST Burow et al. (2002)
activity; slightly reduced
MST activity

Summary table describing the catalytic effects of STR residue mutations. The nature and position of the mutations are indicated as well as the
corresponding AtSTR16 numbering used in Fig. 2. The catalytic cysteine is indicated in bold.

and Yu, 2016). However, another study, focusing on catalytic
cysteine persulfide formation and regeneration by thioredoxin
(TRX) of EcSseA and human MST, reported different results.
In this work, it was found that the step of sulfur transfer be-
tween 3-MP and the thiol group of the catalytic cysteine, and
the protonation state of 3-MP at various pH values is not af-
fected in the corresponding S239A variant (Lec et al., 2018).
These contradictory results raise questions about the import-
ance of the residue found at the +2 position within the active
site loop of MST-type STRs.
Regarding TST-type proteins, the residue at this position is
important for their activity but again effects in protein variants
are variable and sometimes opposite. The change of Lys250 to
Ala in bovine Rhobov abolishes its activity towards thiosul-
fate whereas it does not affect activity towards more complex
substrates such as thiosulfonate (Table 2; Luo and Horowitz,
1994), pointing to substrate-dependent effects. A  mutational
analysis was also performed using A. vinelandii RhdA, an atyp-
ical enzyme that uses only thiosulfate as a sulfur donor in vitro.
Contrary to expectation, the change of Thr232 to Lys or Ala
leads in fact to an increase of TST activity (Pagani et al., 2000).
Finally, the importance of the residue at position +5 (Arg85
in Fig. 2) for protein activity was also investigated. Most STRs
have an Arg or sometimes a Thr but AtSTR13 and human
MOCS3 have an Asp, which makes a considerable difference.
In human MOCS3, the change of this Asp417 to an Arg or a
Thr results in a strong increase of protein activity towards thio-
sulfate (Table 2; Krepinsky and Leimkühler, 2007). Hence, the
fact that AtSTR13 and MOCS3 possess an N-terminal MoeB
domain required for Moco synthesis and thus have particular
substrates/partners may explain why the basic residue usually
found at this position and which seems important for the pro-
tein activity with thiosulfate has been replaced by an acidic
residue.
Dual functions of STR13 in the delivery of
sulfur atoms
Implication in Moco synthesis
Moco consists of a molybdenum atom covalently bound
to two sulfur atoms of a tricyclic pterin moiety referred to
4146 | Selles et al.

as molybdopterin (MPT). Moco is found at the active site
of molybdenum enzymes. In plants, the Moco-dependent
enzymes nitrate reductase (NR), aldehyde oxidase (AO),
xanthine dehydrogenase (XDH), and sulfite oxidase (SO)
function respectively in nitrogen assimilation, abscisic acid
synthesis, purine catabolism, and sulfite detoxification
(Schwarz and Mendel, 2006). Moco biosynthesis is there-
fore crucial and is now well understood (Fig. 4A). The first
step, consisting of the condensation of GTP into cyclic
pyranopterin monophosphate (cPMP), is catalysed by the
Fe–S protein CNX2 and by CNX3 and takes place in the
mitochondrial matrix. cPMP is then exported to the cytosol.
Initially it was suggested that the ABC transporter ATM3
transports cPMP, as the corresponding mutants show a
Moco into apoproteins. In NR and SO, a sulfur atom of a
conserved protein cysteine residue covalently binds the Mo
atom. On the contrary in XDH and AO, the Moco sulfur-
ation is independent of protein cysteine residues and de-
pends on the activity of the cytosolic cysteine desulfurase
ABA3 (Bittner et al., 2001). This CD is relatively atypical as
it consists of two domains, an N-terminal CD domain that
transfers the sulfur atom to a Moco bound to the C-terminal
domain, prior to the transfer and insertion of this sulfurated
Moco into AO and XDH (Wollers et al., 2008). A functional
characterization demonstrated that an Arabidopsis cnx5/str13
mutant containing a change of the conserved Ser149 residue
to Phe accumulates high levels of cPMP, whereas cnx5 null
mutants have a strong growth defect and are sterile (Nakai

Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023

strong phenotype for Moco-containing proteins such as NR
(Bernard et al., 2009; Teschner et al., 2010). Nevertheless, it
was shown that atm3 mutants contain a lower content of
cPMP but a more oxidized mitochondrial matrix, suggesting
that the role of ATM3 in the Moco biosynthesis pathway
may be indirect (Kruse et al., 2018). In the cytosol, two sulfur
atoms are inserted into cPMP by the concerted action of
CNX5, CNX6, and CNX7, resulting in the formation of
MPT (Kaufholdt et al., 2013). Thereafter, CNX1 inserts the
molybdenum atom, which is coordinated by a dithiolene
group, leading to the formation of Moco (Fig. 4A). Finally,
additional modifications occur before the final insertion of
et al., 2012; Kruse et al., 2018). STR13, also named CNX5/
MOCS3, has the particularity of containing a MoeB-like
domain necessary for MPT biosynthesis in the N-terminal
region associated to a Rhd domain with a CRRGND ac-
tive site motif. In vitro assays performed with the human
ortholog, MOCS3, revealed that the persulfidated Rhd do-
main of MOCS3 is able to provide sulfur to MOCS2A, the
human ortholog of CNX7, enabling the synthesis of MPT
(Matthies et al., 2004). Moreover, mutating the catalytic cyst-
eine of the Rhd domain prevents the formation of MPT
revealing the importance of this residue for STR13/CNX5/
MOCS3 function (Matthies et al., 2004).

Moco biosynthesis A Thio-modification of B

cytosolic tRNAs
mitochondrion
CNX2
CNX3
GTP cPMP

tRNA

S uridine
D
cytosol

cPMP S
S CNX5
Cu
CNX7 URM11
SSH
CNX6

MPT
MoO42-
Mg-ATP S
Cu
CNX1
AMP

Moco
2-thiouridine

Fig. 4.  Role of STR13/CNX5 in molybdenum cofactor biosynthesis and thio-modification of cytosolic tRNAs. (A) Simplified representation of molybdenum
cofactor (Moco) biosynthesis in Arabidopsis modified from Kaufholdt et al. (2013). The biosynthesis of Moco starts in the mitochondrial matrix where
guanosine-5′-triphosphate (GTP) is converted to cyclic pyranopterin monophosphate (cPMP) by CNX2 (cofactor for nitrate reductase and xanthine
dehydrogenase 2) and CNX3. The cPMP is then exported to the cytosol and a heteromeric complex, consisting of CNX6 and CNX7, inserts two sulfur
atoms into cPMP and converts it to molybdopterin (MPT). CNX7 receives the sulfur atoms from STR13/CNX5 but the primary sulfur donor of CNX5 (X-S)
is yet unknown. During the last step, one molybdenum atom is inserted via the activity of CNX1, finally leading to the mature Moco. (B) Synthesis of
thiolated nucleoside in tRNAs. STR13/CNX5 provides sulfur to its partner, URM11 (ubiquitin-related modifier 11) and possibly URM12 (not represented),
which is essential for the thio-modification (such as s2U34) of cytosolic tRNAs (Nakai et al., 2012).
 Roles of sulfurtransferases in sulfur trafficking in plants  |  4147
Implication in thio-modification of tRNAs
In addition to its role in Moco biosynthesis, STR13 is also
involved in the thio-modification of cytosolic tRNAs (Fig.
4B; Nakai et  al., 2012). The thio-modification of tRNAs is
one out of ca a hundred possible post-transcriptional modi-
fications that can influence tRNA folding or stability that
would consequently affect translation and ultimately growth
(Phizicky and Hopper, 2010). The 2-thiouridine s2U corres-
ponds to the 2-thio modification that is universally found at
U34 of tRNA species. In yeast and higher eukaryotes, U34 in
tRNALys (anticodon UUU), tRNAGln (anticodon UUG) and
tRNAGlu (anticodon UUC) are modified and sulfurated to
form 5-methyl-2-thiouridine derivatives (xm5s2U), the most
important being the 5-methoxycarbonylmethyl-2-thiouridine
(mcm5s2U) modification in eukaryotic cytosolic tRNAs
(Ranjan and Rodnina, 2017). In eukaryotes, the biosynthesis
of s2U in cytosolic tRNAs involves a protein thiocarboxylate
as intermediate sulfur donor that is functionally and evolu-
tionarily related to the ubiquitin-like post-translational modi-
fication system as demonstrated notably in yeast (Nakai et al.,
2008). Hence, in yeast, both Urm1 and the yeast STR13
ortholog Uba4 are involved in the thio-modification of cyto-
solic tRNAs. Consistently, cytosolic tRNAs from Δurm1 and
Δuba4 cells are not thiolated (Leidel et al., 2009). Arabidopsis
STR13/CNX5 can fully complement the yeast Δuba4 mu-
tant. Both cnx5-1 and cnx5-2 mutants exhibit a dwarf pheno-
type with slightly green and morphologically aberrant leaves
and in addition, the thio-modification of cytosolic tRNAs is
impaired (Nakai et  al., 2012). Interestingly, Arabidopsis pos-
sesses two orthologs of ScUrm1, URM11 and URM12, both
of them capable of complementing the yeast Δurm1 mutant
(Nakai et  al., 2012). As URM11 is the predominantly ex-
pressed isoform, it was suggested that this protein is primarily
responsible for the thio-modification of cytosolic tRNAs.
Nevertheless, an Arabidopsis urm11-1 null mutant did not
exhibit any visible phenotypic changes, although the mutant
lacked thio-modified cytosolic tRNAs. Hence, the strong/le-
thal phenotype observed for str13/cnx5 mutants should be re-
lated to a defect in Moco biosynthesis, not in cytosolic tRNA
modifications (Nakai et  al., 2012). Altogether, these data re-
veal that STR13 represents an essential link between these two
sulfur-dependent biosynthetic pathways but its sulfur donor is
yet unidentified (Fig. 4). In human cells, a cytosolic form of
the cysteine desulfurase NFS1 is proposed to provide sulfur to
the MOCS3 ortholog (Marelja et al., 2008, 2013). In plants, the
sole CD isoform in the cytosol is ABA3. However, aba3 mu-
tants (aba3-1, aba3-2, los5-1, los5-2) have less severe phenotypes
(Xiong et al., 2001; Zhong et al., 2010) than cnx5/str13 mutants
indicating that it should not provide sulfur to STR13, leaving
the question of the sulfur source open.
Roles of STR5 and STR17a in arsenate
tolerance
The presence of inorganic arsenic in soils and fresh water
is a widely recognized problem. Some plant species, such as
the ferns Pteris vittata and Pityrogamma calomelanos, are known
to accumulate arsenic, and thus represent good candidates
for arsenic phytoremediation approaches (Peer et  al., 2006).
Meanwhile, arsenic accumulation in crops is a challenging issue
and particularly in rice, which is the staple diet for more than
half of the world’s population (Williams et al., 2006). Hence,
the investigation of how plants cope with this useless metal is
a decisive research area and recent advances have been made
in Arabidopsis, implicating two STR isoforms, AtSTR5 and
AtSTR17a.
According to the involvement of dual-specificity CDC25
phosphatases in cell cycle regulation, AtSTR5 exhibits phos-
phatase activity in vitro and enhances cyclin-dependent serine/
threonine kinase activity of purified CDK complexes via tyro-
sine dephosphorylation (Landrieu et  al., 2004a,b). Also, the
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
heterologous expression of AtSTR5 in fission yeast reduces the
size of mitotic cells and Arabidopsis str5 mutants are hyper-
sensitive to a replication blocking agent (Sorrell et  al., 2005;
Spadafora et al., 2011). In parallel, other studies have suggested
the involvement of AtSTR5 in arsenate reduction. Interestingly,
both tyrosine phosphatase and arsenate reductase activities
depend on the presence of the catalytic cysteine present in
the active site motif (HCX5R) of the Rhd domain (Landrieu
et  al., 2004a). Plant STR5 genes complement the arsenate-
sensitive phenotype of the E. coli arsC and yeast acr2 mutants
and recombinant proteins display arsenate reductase activity
in vitro (Dhankher et al., 2006; Ellis et al., 2006). Accordingly,
Arabidopsis STR5 down-regulated lines are sensitive to ar-
senate treatments and present a hyperaccumulation of arsenic
when they are cultured in vitro for 3 weeks (Dhankher et al.,
2006). Moreover, a 2-fold higher arsenate reductase activity
was observed in AtSTR5-overexpressing lines in comparison
with wild-type plants (Dissmeyer et  al., 2009). Puzzlingly, in
another study the content of arsenic species was not modified
in both Arabidopsis STR5 knockout or overexpression lines
exposed to arsenate for 0.5, 2, 6 or 24  h as compared with
wild-type plants (Liu et  al., 2012). Another piece of positive
evidence is, however, that transgenic tobacco plants expressing
AtSTR5 are more tolerant to arsenic than wild-type plants
(Nahar et al., 2017). The subcellular localization of AtSTR5 is
also subject to some uncertainty. An experiment using the first
100 amino acids of AtSTR5 fused to green fluorescent pro-
tein (GFP) demonstrated a mitochondrial localization (Narsai
et al., 2011). This might correlate with the N-terminal trun-
cation at the position of the Ser20 reported in an N-terminal
proteome study performed with Arabidopsis seedlings (Venne
et  al., 2015). Nonetheless, this mitochondrial localization of
AtSTR5 would not be compatible with its described func-
tion in cell cycle regulation, which would require a nuclear
localization. More experimental evidence is required here
to validate whether AtSTR5 is indeed expressed in multiple
subcellular compartments.
The involvement of AtSTR17a in arsenate resistance/ac-
cumulation has been independently identified by two groups
performing genome-wide association studies focusing either
on root growth recovery after arsenate exposure in a collection
of 82 Arabidopsis accessions (Sánchez-Bermejo et al., 2014) or
on arsenate accumulation in leaves of 349 Arabidopsis acces-
sions (Chao et al., 2014). Based on expression profile analysis,
4148 | Selles et al.
AtSTR17a is highly expressed in roots (Fig. 1). Experiments
using AtSTR17a–GFP fusion reporter revealed a nucleo-
cytosolic localization (Sánchez-Bermejo et al., 2014) and an ac-
cumulation in root hairs, epidermis, and pericycle cells (Chao
et al., 2014). As AtSTR17a displays an arsenate reductase ac-
tivity and Arabidopsis str17a mutants significantly accumulate
arsenic in leaves, Chao and co-workers proposed that arsenate
reduction by AtSTR17a in roots and notably in the pericycle
may limit arsenic loading into the xylem, avoiding its transfer
to the shoots. The difference in the V[G/A]CXXGXR active
site motif found in AtSTR17a compared with the canon-
ical HCX5R active site encountered in yeast and Arabidopsis
ACR2 led to naming AtSTR17a as ARQ1 (arsenate reduc-
tase QTL1) or HAC1 (high arsenic content 1)  (Chao et  al.,
2014; Sánchez-Bermejo et al., 2014). Both groups also sought
to establish a link between AtSTR5 and AtSTR17a in arsenate
tolerance by investigating the impact of AtSTR5/ACR2 mu-
tation on their respective experimental schemes. No significant
changes of the root growth recovery or of arsenate leaf con-
tent were observed in a str5 knockout mutant and the analysis
of a double str17a str5 mutant did not highlight an epistasic
interaction between the two genes (Chao et al., 2014; Sánchez-
Bermejo et  al., 2014). Altogether, these observations support
a role of both proteins in arsenate tolerance but raise some
questions about their respective roles, possibly pointing to their
involvement in distinct mechanisms/pathways.
From a molecular point of view, it is informative to correlate
the polymorphism present in naturally variable sequences to
some phenotypes, as was done for AtSTR17a between the
Col-0 ecotype and two arsenate hypersensitive ecotypes, Kas-1
and Kr-0. In Kr-0, the AtSTR17a gene exhibits a premature
stop codon leading to the synthesis of an inactive protein de-
void of the Rhd domain (Chao et  al., 2014). In Kas-1, the
AtSTR17a protein has differences in the N-terminal part but
also four internal substitutions in the Rhd domain. A chimeric
AtSTR17a Kas-1 protein bearing the N-terminal extension
and the internal substitutions found in the Col-0 allele is able
to complement the arsC mutant. Concomitantly, mutagenesis
analysis of the AtSTR17a Col-0 protein bearing a combination
of three substitutions from the AtSTR17a Kas-1 protein dem-
onstrates a non-additive effect of these mutations, but specific
substitution pairs needed for loss of activity (Sánchez-Bermejo
et  al., 2014). Altogether, these data illustrate the natural vari-
ability in the AtSTR17a-related arsenate resistance/accumu-
lation encountered in different Arabidopsis accessions and the
complexity of structural features that can promote or desta-
bilize enzymatic reactions catalysed by STR members.
Multiple roles for the mitochondrial STR1
In plants, three pathways for cysteine degradation have been
described (for a detailed review see Hildebrandt et al., 2015),
the mitochondrial STR1 contributing to one of these. This
pathway starts with the transamination of cysteine to 3-MP, a
reaction catalysed by a so far unknown aminotransferase (Fig.
5A). STR1 forms pyruvate from 3-MP, the persulfide group
concomitantly formed on the catalytic cysteine being eventually
transferred to GSH. The resulting GSSH could then be oxi-
dized to sulfite (SO32−) by the sulfur dioxygenase ethylmalonic
encephalopathy protein 1 (ETHE1) (Krüßel et  al., 2014).
Sulfite is likely preferentially transferred to peroxisomes and
oxidized to sulfate by SO (Brychkova et al., 2007, 2013; Lang
et  al., 2007), but it could eventually be further converted to
thiosulfate (S2O32−) via a persulfidated STR1 (Fig. 5B; Krüßel
et al., 2014; Höfler et al., 2016). A role of STR1 in sulfite de-
toxification was suggested by the observation of a higher thio-
sulfate level and STR-dependent sulfite-consuming activity in
RNAi-silenced plants for SO in comparison with wild-type
plants (Brychkova et  al., 2013). Instead of being transferred
to GSH, the persulfide group formed on the catalytic Cys of
STR1 could be transmitted to other sulfur acceptor molecules
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
or could lead to the production of hydrogen sulfide (H2S)
by reaction with mitochondrial disulfide reductases, notably
TRXo1 and TRXo2 (Fig. 5B). This aspect is further discussed
in the next section. An alternative route by which STR1 could
obtain sulfur atoms and deliver them to target compounds/
proteins without relying on 3-MP (Fig. 5) is via the mitochon-
drial cysteine desulfurase NFS1 because an interaction with
the MST TUM1 has been reported in human and yeast (Noma
et al., 2009; Fräsdorf et al., 2014).This enzyme catalyses cysteine
desulfuration into alanine, providing the sulfur required for de
novo Fe–S cluster biogenesis (Couturier et  al., 2013). In this
pathway, NFS1 forms an assembly complex with several other
proteins and the persulfide group formed on its active site cyst-
eine residue provides the sulfur atoms used for building Fe–S
clusters. It is worth noting that a similar reaction is catalysed by
NFS2 in chloroplasts (Couturier et  al., 2013). Whether plant
NFS1 and NFS2 do interact with some STRs is not known.
So far, STR1 is the sole MST identified in Arabidopsis
mitochondria (Nakamura et  al., 2000; Papenbrock and
Schmidt, 2000; Senkler et al., 2016). The second MST isoform,
named STR2 and displaying a high sequence identity with
STR1 (79%), was shown to be localized in the cytosol using
immunoblots on fractionated cellular extracts and GFP fu-
sion (Nakamura et al., 2000; Papenbrock and Schmidt, 2000).
However, a longer STR2 transcript (At1g16460.2) is also ex-
pressed and encodes a protein with an N-terminal extension
of 24 residues that is predicted to constitute a mitochondrial
targeting sequence. Hence, this will have to be experimentally
addressed even though STR2 was not detected in a recent
mitochondrial proteome analysis, unlike STR1 (Senkler et al.,
2016). The study of str mutants indicates that STR1 accounts
for 50–80% of the MST activity whereas str2 mutant seedlings
have a wild-type MST activity level (Mao et al., 2011; Höfler
et al., 2016). However, their physiological and molecular roles
in planta have not been elucidated because str2 null mutants do
not have any phenotype and str1 null mutants show a shrunken
seed phenotype with ~87.5% of the embryos arrested at the
heart stage (Mao et al., 2011). The residual 12.5% can develop
further and show a normal development of vegetative tissue.
Importantly, the double str1 str2 mutant is not viable indicating
that both STRs ensure an essential function at least for embryo
development, but that STR1 likely has a predominant role
(Mao et al., 2011). Similar to STR1, ETHE1 null mutants show
an embryo-lethal phenotype (Krüßel et al., 2014).Whereas this
 Roles of sulfurtransferases in sulfur trafficking in plants  |  4149

Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023

Fig. 5.  Confirmed or hypothetical roles of STR1 in sulfur trafficking in mitochondria. (A) Persulfidation of mitochondrial STR1. This might occur upon
reaction with 3-mercaptopyruvate (3MP) formed from cysteine by a so far unknown cysteine aminotransferase (CAT) or via the transfer of a persulfide
formed on the NFS1 cysteine desulfurase. Whether the latter pathway operates in plants in unknown but the human and yeast STR and NFS orthologs
were shown to interact. (B) Putative reaction pathways of persulfidated STR1. (1) A reaction with reduced glutathione (GSH) generates glutathione
persulfide (GSSH) that can then either react with another GSH molecule leading to the release of hydrogen sulfide (H2S) and oxidized glutathione (GSSG)
or be used as substrate by the sulfur dioxygenase ethylmalonic encephalopathy protein 1 (ETHE1) to form sulfite (SO32−). Sulfite is either detoxified
in sulfate by sulfite oxidase after its transport in peroxisomes, or converted to thiosulfate in mitochondria by persulfidated STR1. (2) A reaction with
thioredoxin o (TRXo) isoforms would lead to H2S formation. (3) In addition to the already described reaction with GSH, TRXo, or sulfite, a reaction with
mitochondrial protein or non-protein acceptors such as cysteine results in the formation persulfidated proteins or of low molecular mass per/polysulfides,
respectively.

would fit with their functional association described in animals
(Kabil and Banerjee, 2014) and with the existence of naturally
occurring chimeric proteins containing a sulfur dioxygenase
domain fused to one or several Rhd domains (Shen et al., 2015;
Motl et  al., 2017), the mutants have substantial differences.
A  knockdown mutant of ETHE1 (ethe1-1 line) with 1% of
wild-type sulfur dioxygenase activity produces seeds morpho-
logically indistinguishable from the wild-type although with a
delayed embryo development. Moreover, germination, vegeta-
tive growth under short day conditions, and senescence are dif-
ferentially impacted in the ethe1-1 and str1-1 mutants (Krüßel
et al., 2014; Höfler et al., 2016).
Hence, it might be that STR1/2 possess additional physio-
logical functions in this compartment. Mao and colleagues
speculated about a role of both STRs in the protection of
developing embryos by converting cyanide, a co-product of
ethylene biosynthesis, to the less toxic thiocyanate (Mao et al.,
2011) as proposed previously for the mitochondrial MST from
rat to protect cytochrome c oxidase against cyanide (Nagahara
et al., 1999). However, a direct role in cyanide detoxification
seems unlikely because enzymes catalysing this reaction use
thiosulfate, which is not a good substrate for STR1/2, that ra-
ther produce thiosulfate (Höfler et al., 2016). Hence, an indirect
contribution is possible if STR1/2 produce thiosulfate that is
used by other mitochondrial STRs, such as STR10. Still, mito-
chondria already contain the β-cyanoalanine synthase CAS-
C1 that is involved in the detoxification of cyanide through
the formation of β-cyanoalanine (García et  al., 2010). For
comparison purpose, it is worth noting that the cas-c1 mutant
accumulates cyanide notably in roots, strongly inhibiting root
hair elongation and altering the plant response to pathogens
(García et al., 2013). In the absence of additional data, for in-
stance about cyanide, sulfite, and thiosulfate levels in both str1
and str2 mutants, a contribution of STR1/2 in cyanide detoxi-
fication remains uncertain.
Production of hydrogen sulfide by STRs
The study of H2S production and its associated physiological
and signaling roles in plants but also in other organisms is a
recently emerging area, many aspects remaining mysterious,
notably in plants (Mustafa et al., 2009; Paul and Snyder, 2012;
Greiner et  al., 2013; Longen et  al., 2016; Aroca et  al., 2017).
In plants, H2S is linked to the regulation of many develop-
mental stages and to stress response mechanisms, being pos-
sibly produced in several subcellular compartments (Hancock
and Whiteman, 2014). Noticeably, it is produced in chloro-
plasts in the frame of the reductive assimilation of sulfate
(Takahashi et  al., 2011), in mitochondria through the syn-
thesis of β-cyanoalanine by CAS-C1 (Álvarez et  al., 2012)
and in the cytosol through cysteine degradation by L-cysteine
desulfhydrase 1 (DES1) (Álvarez et al., 2010). In mammals, H2S
is produced by the side-reactions of cystathionine β-synthase
and cystathionine γ-lyase, two enzymes of the trans-sulfuration
pathway that normally convert cysteine to homocysteine (re-
viewed in Kabil and Banerjee, 2014). In a third pathway, H2S
is released when persulfide-bound forms of the mitochondrial
MST interact with a reductant such as TRX or dihydrolipoic
acid as shown in mice (Shibuya et  al., 2009; Mikami et  al.,
2011;Yadav et al., 2013).The released H2S may serve for modi-
fying specific proteins (see below). Interestingly, in mouse
4150 | Selles et al.
hepatocytes lacking both TRX reductase and glutathione re-
ductase the level of persulfidated proteins is markedly elevated
(Dóka et al., 2016). Although this underlines the importance
of both reducing systems for protein persulfidation, their exact
contribution remains unclear. It may simply be that they serve
for the reduction of persulfidated proteins.
A connection between STR1 and the reducing systems seems
true in plants as well given that STR1 has been repeatedly iso-
lated by pull-down approaches as a partner of TRX including
the mitochondrial TRXo1 isoform from Arabidopsis (Balmer
et al., 2004;Yoshida et al., 2013; Pérez-Pérez et al., 2017). In this
species, the STR1–TRXo1 interaction has been confirmed
in vivo using bimolecular fluorescence complementation ex-
periments (Henne et  al., 2015). Interestingly, in vivo inter-
actions between TSTs, such as STR14, STR15, and STR16,
and chloroplastic TRXs have been reported suggesting that
the STR–TRX interaction is not restricted to MSTs. These
STRs may be also involved in H2S biogenesis in addition to
their capacity to perform trans-persulfidation reactions. An
STR–GRX interaction has not been reported so far but nat-
ural fusion proteins containing both protein domains exist in
some prokaryotes. If such an interaction is confirmed, it will be
important to make a distinction between the TRX and GSH/
GRX reducing systems (Fig. 6). Indeed, the majority of plant
TRXs employ two cysteines to reduce disulfide substrates
whereas GRXs mostly use a single cysteine (even though
there is another cysteine in the active site signature). Hence,
the presence of a persulfide group on a catalytic cysteine of a
TRX should be short-lived because of the presence of a re-
solving cysteine and should lead to the release of H2S. On the
contrary, a persulfide group formed on the catalytic cysteine
of a GRX may be more stable and allow trans-persulfidation
Fig. 6.  Possible interaction of STRs with the cellular reducing components. The reaction of persulfidated STRs with dithiol thioredoxin (TRX) or
glutaredoxin (GRX) isoforms generates H2S (1) whereas reaction with monothiol proteins/molecules (monothiol GRX, GSH, cysteine) generates
persulfidated molecules that could participate to subsequent trans-persulfidation reactions (2).
reactions with some partner proteins (Mishanina et al., 2015).
Accordingly, seven GRX isoforms and 16 TRX isoforms were
retrieved as persulfidated proteins in Arabidopsis (Aroca et al.,
2017). Such a relay with TRX or GRX may be convenient to
achieve an additional degree of specificity towards target pro-
teins (Fig. 6).
The signaling functions of H2S likely rely on the reactions
with specific proteins leading to their persulfidation. It is im-
portant to mention that H2S will react only with oxidized
cysteines, possibly sulfenylated, nitrosylated, glutathionylated,
or forming a disulfide bond. Hence, a complex interplay
exists between all these redox post-translational modifica-
tions. Alternatively, H2S-mediated protein persulfidation oc-
curs when polysulfides (RS(S)nH and RS(S)nSR) derived from
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
H2S react with cysteine thiol or thiolate (Greiner et al., 2013).
Remarkable amounts of these polysulfides have been detected
in vivo in mammalian cells in addition to cysteine-persulfide
(Cys-SSH) and glutathione-persulfide (GSSH) (Ida et  al.,
2014). All persulfide species present in cells, i.e. polysulfides
(notably H2S2 and H2S3), low molecular mass persulfides (Cys-
SSH and GSSH) and persulfidated protein cysteine residues,
have been designated as forming the bound sulfane sulfur
forms. The levels of total persulfidated species in the brain of
MST-KO mice are less than 50% of that in the brain of wild-
type mice indicating that mitochondrial MST is indeed an im-
portant actor (Kimura et al., 2015, 2017). Similarly, E. coli SseA
is involved in the production of reactive sulfane sulfur and not-
ably GSSH and GSSSH (Li et al., 2019).
Whether plant MST orthologs, STR1 and STR2, partici-
pate in the production of such molecules in plants is poorly
documented. Nevertheless, in addition to its interaction with
TRXo1 leading to H2S release, Arabidopsis STR1 is able to
 Roles of sulfurtransferases in sulfur trafficking in plants  |  4151
produce GSSH (Höfler et  al., 2016). Moreover, a proteomic
analysis identified 2015 persulfidated proteins from leaves of
wild-type Arabidopsis plants (Aroca et  al., 2017). A  similar
number (2130 persulfidated proteins) and almost the same
proteins (85% of concordance) were identified in the des1 mu-
tant, which is defective for a cytosolic protein generating H2S
(Aroca et al., 2017). This suggests that the major factors pro-
moting protein persulfidation may indeed be STR1 and STR2
but also other STRs or yet uncharacterized proteins. In mam-
mals, the ubiquitous mitochondrial cysteinyl-tRNA synthetase
has been characterized as the major Cys-SSH synthase in vivo,
playing an important role in the formation of both low mo-
lecular mass polysulfides and persulfidated proteins (Akaike
et al., 2017). To conclude, elucidating the molecular mechan-
isms underlying H2S-mediated protein persulfidation and its
regulation in plants should become a priority and more intense
investigations are necessary to understand the effect of these
modifications on protein activity and function.
Conclusions
Despite the physiological importance of many sulfur-containing
molecules, the molecular mechanisms associated with the traf-
ficking and specific delivery of sulfur atoms to their acceptors
remain unclear in a number of pathways. Through their ability
to act as sulfur carrier proteins, STRs fulfill important and di-
verse roles in these reactions in all organisms and more particu-
larly in plants. Accordingly, the STR family in higher plants is
composed on average of 20 members that are quite divergent
considering their primary sequence, protein domain organiza-
tion, and subcellular localization. The best documented func-
tions for STR members include roles of STR5 and STR17a
in arsenate tolerance, of STR13/CNX5 in Moco biosynthesis
and thio-modification of cytosolic tRNAs, and of STR1 in
cysteine degradation and H2S formation.
Surprisingly, the exact roles of other STRs and more par-
ticularly those localized in chloroplasts are yet to be explored.
This is all the more important as these isoforms represent
nearly half of all STRs and sulfide and sulfite are produced in
plastids during sulfate reduction. STR4/thylakoid rhodanese-
like protein (TROL), which is inactive as STR, is required for
tethering ferredoxin-NADP oxidoreductase to the thylakoid
membrane (Jurić et al., 2009). This interaction is proposed to
control the flow of photosynthetic electrons prior to activation
of pseudo-cyclic electron flow (Vojta et al., 2015).
In addition to genetic and physiological studies, experi-
ments aiming at identifying STR partners are required and
should help in defining their precise roles and eventually lead
to the identification of new STR-dependent pathways. In vitro
analyses of the interactions between STRs and sulfur donors/
acceptors appear crucial and this necessitates more systematic
exploration of the biochemical and structural properties of
STR isoforms alone or upon interaction.
Whether and how STR activity is regulated is another as-
pect to investigate in the future. As the role of STRs in sulfur
exchange reactions relies on the presence of a catalytic cyst-
eine in their Rhd domain, any modifications of this mandatory
residue will represent a way for their post-translational regula-
tion. Although limited evidence exist so far, several Arabidopsis
STR isoforms were identified as possessing reversibly oxi-
dized cysteines (Slade et al., 2015) and STR1 was identified as
sulfenylated in plants treated with H2O2 (Akter et al., 2015; De
Smet et al., 2019). This may further link STRs with the TRX
and GRX reducing systems besides their connection for the
formation of H2S and for protein (trans)-persulfidation.
Acknowledgements
The salary of AM was funded by grants of the French National Research
Agency (ANR) as part of the ‘Investissements d’Avenir’ program (ANR-
11-LABX-0002-01, Lab of Excellence ARBRE) and of the SULPAR
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
(17_GE2_016) GrandEst region contract. AM also greatly appreciates a
postdoc fellowship from the Alexander von Humboldt Foundation. The
salary of BS and the work on plant sulfurtransferases are supported by
grant of the ANR-16-CE20-0012 contract.
References
Akaike  T, Ida  T, Wei  FY, et  al. 2017. Cysteinyl-tRNA synthetase gov-
erns cysteine polysulfidation and mitochondrial bioenergetics. Nature
Communications 8, 1177.
Akter  S, Huang  J, Bodra  N, et  al. 2015. DYn-2 based identification of
Arabidopsis sulfenomes. Molecular & Cellular Proteomics 14, 1183–1200.
Alphey  MS, Williams  RA, Mottram  JC, Coombs  GH, Hunter  WN.
2003. The crystal structure of Leishmania major 3-mercaptopyruvate
sulfurtransferase. A  three-domain architecture with a serine protease-
like triad at the active site. The Journal of Biological Chemistry 278,
48219–48227.
Álvarez  C, Calo  L, Romero  LC, García  I, Gotor  C. 2010. An
O-acetylserine(thiol)lyase homolog with L-cysteine desulfhydrase activity
regulates cysteine homeostasis in Arabidopsis. Plant Physiology 152,
656–669.
Álvarez C, García I, Romero LC, Gotor C. 2012. Mitochondrial sulfide
detoxification requires a functional isoform O-acetylserine(thiol)lyase C in
Arabidopsis thaliana. Molecular Plant 5, 1217–1226.
Aroca  A, Benito  JM, Gotor  C, Romero  LC. 2017. Persulfidation
proteome reveals the regulation of protein function by hydrogen sulfide in
diverse biological processes in Arabidopsis. Journal of Experimental Botany
68, 4915–4927.
Babicki  S, Arndt  D, Marcu  A, Liang  Y, Grant  JR, Maciejewski  A,
Wishart  DS. 2016. Heatmapper: web-enabled heat mapping for all.
Nucleic Acids Research 44, W147–W153.
Balmer Y, Vensel WH, Tanaka CK, et al. 2004. Thioredoxin links redox to
the regulation of fundamental processes of plant mitochondria. Proceedings
of the National Academy of Sciences, USA 101, 2642–2647.
Bartels A. 2006. Functional characterisation of sulfurtransferase proteins in
higher plants. PhD thesis, Leibniz Universität Hannover.
Bartels  A, Mock  HP, Papenbrock  J. 2007. Differential expression of
Arabidopsis sulfurtransferases under various growth conditions. Plant
Physiology and Biochemistry 45, 178–187.
Bauer  M, Dietrich  C, Nowak  K, Sierralta  WD, Papenbrock  J. 2004.
Intracellular localization of Arabidopsis sulfurtransferases. Plant Physiology
135, 916–926.
Behshad E, Bollinger JM Jr. 2009. Kinetic analysis of cysteine desulfurase
CD0387 from Synechocystis sp. PCC 6803: formation of the persulfide
intermediate. Biochemistry 48, 12014–12023.
Bernard DG, Cheng Y, Zhao Y, Balk J. 2009. An allelic mutant series of
ATM3 reveals its key role in the biogenesis of cytosolic iron-sulfur proteins in
Arabidopsis. Plant Physiology 151, 590–602.
Bittner F, Oreb M, Mendel RR. 2001. ABA3 is a molybdenum cofactor
sulfurase required for activation of aldehyde oxidase and xanthine dehydro-
genase in Arabidopsis thaliana. The Journal of Biological Chemistry 276,
40381–40384.
4152 | Selles et al.
Blumenthal KM, Heinrikson RL. 1971. Structural studies of bovine liver
rhodanese. I. Isolation and characterization of two active forms of the en-
zyme. The Journal of Biological Chemistry 246, 2430–2437.
Bordo D, Bork P. 2002. The rhodanese/Cdc25 phosphatase superfamily.
Sequence-structure-function relations. EMBO Reports 3, 741–746.
Brychkova G, Grishkevich V, Fluhr R, Sagi M. 2013. An essential role for
tomato sulfite oxidase and enzymes of the sulfite network in maintaining leaf
sulfite homeostasis. Plant Physiology 161, 148–164.
Brychkova G, Xia Z, Yang G, Yesbergenova Z, Zhang Z, Davydov O,
Fluhr  R, Sagi  M. 2007. Sulfite oxidase protects plants against sulfur di-
oxide toxicity. The Plant Journal 50, 696–709.
Burow  M, Kessler  D, Papenbrock  J. 2002. Enzymatic activity of the
Arabidopsis sulfurtransferase resides in the C-terminal domain but is
boosted by the N-terminal domain and the linker peptide in the full-length
enzyme. Biological Chemistry 383, 1363–1372.
Chao  DY, Chen  Y, Chen  J, Shi  S, Chen  Z, Wang  C, Danku  JM,
Zhao FJ, Salt DE. 2014. Genome-wide association mapping identifies a
new arsenate reductase enzyme critical for limiting arsenic accumulation in
plants. PLoS Biology 12, e1002009.
Cipollone R, Ascenzi P, Visca P. 2007. Common themes and variations
in the rhodanese superfamily. IUBMB Life 59, 51–59.
Colnaghi  R, Cassinelli  G, Drummond  M, Forlani  F, Pagani  S. 2001.
Properties of the Escherichia coli rhodanese-like protein SseA: contribution
of the active-site residue Ser240 to sulfur donor recognition. FEBS Letters
500, 153–156.
Cornilescu  G, Vinarov  DA, Tyler  EM, Markley  JL, Cornilescu  CC.
2006. Solution structure of a single-domain thiosulfate sulfurtransferase
from Arabidopsis thaliana. Protein Science 15, 2836–2841.
Couturier J, Jacquot JP, Rouhier N. 2013. Toward a refined classification
of class I dithiol glutaredoxins from poplar: biochemical basis for the defin-
ition of two subclasses. Frontiers in Plant Science 4, 518.
De  Smet  B, Willems  P, Fernandez-Fernandez  AD, Alseekh  S,
Fernie  AR, Messens  J, Van  Breusegem  F. 2019. In vivo detec-
tion of protein cysteine sulfenylation in plastids. The Plant Journal 97,
765–778.
Dhankher  OP, Rosen  BP, McKinney  EC, Meagher  RB. 2006.
Hyperaccumulation of arsenic in the shoots of Arabidopsis silenced for
arsenate reductase (ACR2). Proceedings of the National Academy of
Sciences, USA 103, 5413–5418.
Dissmeyer N, Weimer AK, Pusch S, et al. 2009. Control of cell prolifer-
ation, organ growth, and DNA damage response operate independently of
dephosphorylation of the Arabidopsis Cdk1 homolog CDKA;1. The Plant
Cell 21, 3641–3654.
Dóka É, Pader I, Bíró A, et al. 2016. A novel persulfide detection method
reveals protein persulfide- and polysulfide-reducing functions of thioredoxin
and glutathione systems. Science Advances 2, e1500968.
Ellis  DR, Gumaelius  L, Indriolo  E, Pickering  IJ, Banks  JA, Salt  DE.
2006. A novel arsenate reductase from the arsenic hyperaccumulating fern
Pteris vittata. Plant Physiology 141, 1544–1554.
Fräsdorf  B, Radon  C, Leimkühler  S. 2014. Characterization and inter-
action studies of two isoforms of the dual localized 3-mercaptopyruvate
sulfurtransferase TUM1 from humans. The Journal of Biological Chemistry
289, 34543–34556.
Friso  G, Giacomelli  L, Ytterberg  AJ, Peltier  JB, Rudella  A, Sun  Q,
Wijk  KJ. 2004. In-depth analysis of the thylakoid membrane proteome
of Arabidopsis thaliana chloroplasts: new proteins, new functions, and a
plastid proteome database. The Plant Cell 16, 478–499.
García I, Castellano JM, Vioque B, Solano R, Gotor C, Romero LC.
2010. Mitochondrial β-cyanoalanine synthase is essential for root hair for-
mation in Arabidopsis thaliana. The Plant Cell 22, 3268–3279.
García I, Rosas T, Bejarano ER, Gotor C, Romero LC. 2013. Transient
transcriptional regulation of the CYS-C1 gene and cyanide accumulation
upon pathogen infection in the plant immune response. Plant Physiology
162, 2015–2027.
Greiner  R, Pálinkás  Z, Bäsell  K, Becher  D, Antelmann  H, Nagy  P,
Dick TP. 2013. Polysulfides link H2S to protein thiol oxidation. Antioxidants
& Redox Signaling 19, 1749–1765.
Guretzki S, Papenbrock J. 2011. Characterization of the sulfurtransferase
family from Oryza sativa L. Plant Physiology and Biochemistry 49,
1064–1070.
Hancock  JT, Whiteman  M. 2014. Hydrogen sulfide and cell signaling:
team player or referee? Plant Physiology and Biochemistry 78, 37–42.
Hatzfeld  Y, Saito  K. 2000. Evidence for the existence of rhodanese
(thiosulfate:cyanide sulfurtransferase) in plants: preliminary characteriza-
tion of two rhodanese cDNAs from Arabidopsis thaliana. FEBS Letters 470,
147–150.
Henne  M, König  N, Triulzi  T, Baroni  S, Forlani  F, Scheibe  R,
Papenbrock J. 2015. Sulfurtransferase and thioredoxin specifically interact
as demonstrated by bimolecular fluorescence complementation analysis
and biochemical tests. FEBS Open Bio 5, 832–843.
Hildebrandt TM, Grieshaber MK. 2008. Redox regulation of mitochon-
drial sulfide oxidation in the lugworm, Arenicola marina. The Journal of
Experimental Biology 211, 2617–2623.
Hildebrandt  TM, Nunes  Nesi  A, Araújo  WL, Braun  HP. 2015. Amino
acid catabolism in plants. Molecular Plant 8, 1563–1579.
Höfler  S, Lorenz  C, Busch  T, Brinkkötter  M, Tohge  T, Fernie  AR,
Braun  HP, Hildebrandt  TM. 2016. Dealing with the sulfur part of cyst-
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
eine: four enzymatic steps degrade l-cysteine to pyruvate and thiosulfate in
Arabidopsis mitochondria. Physiologia Plantarum 157, 352–366.
Huang  GT, Yu  JS. 2016. Enzyme catalysis that paves the way for
S-sulfhydration via sulfur atom transfer. The Journal of Physical Chemistry.
B 120, 4608–4615.
Ida  T, Sawa  T, Ihara  H, et  al. 2014. Reactive cysteine persulfides and
S-polythiolation regulate oxidative stress and redox signaling. Proceedings
of the National Academy of Sciences, USA 111, 7606–7611.
Jurić  S, Hazler-Pilepić  K, Tomasić  A, et  al. 2009. Tethering of
ferredoxin:NADP+ oxidoreductase to thylakoid membranes is mediated by
novel chloroplast protein TROL. The Plant Journal 60, 783–794.
Kabil  O, Banerjee  R. 2014. Enzymology of H2S biogenesis, decay and
signaling. Antioxidants & Redox Signaling 20, 770–782.
Kaufholdt  D, Gehl  C, Geisler  M, Jeske  O, Voedisch  S, Ratke  C,
Bollhöner  B, Mendel  RR, Hänsch  R. 2013. Visualization and quantifi-
cation of protein interactions in the biosynthetic pathway of molybdenum
cofactor in Arabidopsis thaliana. Journal of Experimental Botany 64,
2005–2016.
Kimura  Y, Koike  S, Shibuya  N, Lefer  D, Ogasawara  Y, Kimura  H.
2017. 3-Mercaptopyruvate sulfurtransferase produces potential redox regu-
lators cysteine- and glutathione-persulfide (Cys-SSH and GSSH) together
with signaling molecules H2S2, H2S3 and H2S. Scientific Reports 7, 10459.
Kimura  Y, Toyofuku  Y, Koike  S, Shibuya  N, Nagahara  N, Lefer  D,
Ogasawara Y, Kimura H. 2015. Identification of H2S3 and H2S produced
by 3-mercaptopyruvate sulfurtransferase in the brain. Scientific Reports 5,
14774.
Klepikova AV, Kasianov AS, Gerasimov ES, Logacheva MD, Penin AA.
2016. A high resolution map of the Arabidopsis thaliana developmental tran-
scriptome based on RNA-seq profiling. The Plant Journal 88, 1058–1070.
Krepinsky K, Leimkühler S. 2007. Site-directed mutagenesis of the active
site loop of the rhodanese-like domain of the human molybdopterin syn-
thase sulfurase MOCS3. Major differences in substrate specificity between
eukaryotic and bacterial homologs. The FEBS Journal 274, 2778–2787.
Kruse I, Maclean AE, Hill L, Balk J. 2018. Genetic dissection of cyclic
pyranopterin monophosphate biosynthesis in plant mitochondria. The
Biochemical Journal 475, 495–509.
Krüßel  L, Junemann  J, Wirtz  M, et  al. 2014. The mitochondrial sulfur
dioxygenase ETHYLMALONIC ENCEPHALOPATHY PROTEIN1 is required
for amino acid catabolism during carbohydrate starvation and embryo de-
velopment in Arabidopsis. Plant Physiology 165, 92–104.
Landrieu  I, da  Costa  M, De  Veylder  L, et  al. 2004a. A small CDC25
dual-specificity tyrosine-phosphatase isoform in Arabidopsis thaliana.
Proceedings of the National Academy of Sciences, USA 101, 13380–13385.
Landrieu I, Hassan S, Sauty M, Dewitte F, Wieruszeski JM, Inzé D,
De  Veylder  L, Lippens  G. 2004b. Characterization of the Arabidopsis
thaliana Arath;CDC25 dual-specificity tyrosine phosphatase. Biochemical
and Biophysical Research Communications 322, 734–739.
Lang  C, Popko  J, Wirtz  M, Hell  R, Herschbach  C, Kreuzwieser  J,
Rennenberg  H, Mendel  RR, Hänsch  R. 2007. Sulphite oxidase as
key enzyme for protecting plants against sulphur dioxide. Plant, Cell &
Environment 30, 447–455.
Lanz ND, Booker SJ. 2015. Auxiliary iron-sulfur cofactors in radical SAM
enzymes. Biochimica et Biophysica Acta 1853, 1316–1334.
 Roles of sulfurtransferases in sulfur trafficking in plants  |  4153
Lec  J-C, Boutserin  S, Mazon  H, Mulliert  G, Boschi-Muller  S,
Talfournier F. 2018. Unraveling the mechanism of cysteine persulfide for-
mation catalyzed by 3-mercaptopyruvate sulfurtransferases. ACS Catalysis
8, 2049–2059.
Leidel  S, Pedrioli  PG, Bucher  T, Brost  R, Costanzo  M, Schmidt  A,
Aebersold  R, Boone  C, Hofmann  K, Peter  M. 2009. Ubiquitin-related
modifier Urm1 acts as a sulphur carrier in thiolation of eukaryotic transfer
RNA. Nature 458, 228–232.
Li K, Xin Y, Xuan G, Zhao R, Liu H, Xia Y, Xun L. 2019. Escherichia coli
uses separate enzymes to produce H2S and reactive sulfane sulfur from
L-cysteine. Frontiers in Microbiology 10, 298.
Libiad  M, Motl  N, Akey  DL, Sakamoto  N, Fearon  ER, Smith  JL,
Banerjee  R. 2018. Thiosulfate sulfurtransferase-like domain-containing 1
protein interacts with thioredoxin. The Journal of Biological Chemistry 293,
2675–2686.
Liu  W, Schat  H, Bliek  M, Chen  Y, McGrath  SP, George  G, Salt  DE,
Zhao FJ. 2012. Knocking out ACR2 does not affect arsenic redox status in
Arabidopsis thaliana: implications for as detoxification and accumulation in
plants. PLoS ONE 7, e42408.
Longen S, Richter F, Köhler Y, Wittig I, Beck KF, Pfeilschifter J. 2016.
Quantitative persulfide site identification (qPerS-SID) reveals protein targets
of H2S releasing donors in mammalian cells. Scientific Reports 6, 29808.
Luo GX, Horowitz PM. 1994. The sulfurtransferase activity and structure
of rhodanese are affected by site-directed replacement of Arg-186 or Lys-
249. The Journal of Biological Chemistry 269, 8220–8225.
Mao  G, Wang  R, Guan  Y, Liu  Y, Zhang  S. 2011. Sulfurtransferases 1
and 2 play essential roles in embryo and seed development in Arabidopsis
thaliana. The Journal of Biological Chemistry 286, 7548–7557.
Marelja  Z, Mullick  Chowdhury  M, Dosche  C, Hille  C, Baumann  O,
Löhmannsröben  HG, Leimkühler  S. 2013. The L-cysteine desulfurase
NFS1 is localized in the cytosol where it provides the sulfur for molybdenum
cofactor biosynthesis in humans. PLoS ONE 8, e60869.
Marelja Z, Stöcklein W, Nimtz M, Leimkühler S. 2008. A novel role for
human Nfs1 in the cytoplasm: Nfs1 acts as a sulfur donor for MOCS3,
a protein involved in molybdenum cofactor biosynthesis. The Journal of
Biological Chemistry 283, 25178–25185.
Matthies A, Rajagopalan KV, Mendel RR, Leimkühler S. 2004. Evidence
for the physiological role of a rhodanese-like protein for the biosynthesis of
the molybdenum cofactor in humans. Proceedings of the National Academy
of Sciences, USA 101, 5946–5951.
Mikami  Y, Shibuya  N, Kimura  Y, Nagahara  N, Ogasawara  Y,
Kimura  H. 2011. Thioredoxin and dihydrolipoic acid are required for
3-mercaptopyruvate sulfurtransferase to produce hydrogen sulfide. The
Biochemical Journal 439, 479–485.
Mishanina  TV, Libiad  M, Banerjee  R. 2015. Biogenesis of reactive
sulfur species for signaling by hydrogen sulfide oxidation pathways. Nature
Chemical Biology 11, 457–464.
Moseler A, Selles B, Rouhier N, Couturier J. 2019. Novel insights into
the diversity of the sulfurtransferase family in photosynthetic organisms with
emphasis on oak. New Phytologist, doi: 10.1111/nph.15870.
Motl  N, Skiba  MA, Kabil  O, Smith  JL, Banerjee  R. 2017. Structural
and biochemical analyses indicate that a bacterial persulfide dioxygenase-
rhodanese fusion protein functions in sulfur assimilation. The Journal of
Biological Chemistry 292, 14026–14038.
Mueller EG. 2006. Trafficking in persulfides: delivering sulfur in biosynthetic
pathways. Nature Chemical Biology 2, 185–194.
Mustafa AK, Gadalla MM, Sen N, Kim S, Mu W, Gazi SK, Barrow RK,
Yang  G, Wang  R, Snyder  SH. 2009. H2S signals through protein
S-sulfhydration. Science Signaling 2, ra72.
Nagahara N, Ito T, Minami M. 1999. Mercaptopyruvate sulfurtransferase
as a defense against cyanide toxication: molecular properties and mode of
detoxification. Histology and Histopathology 14, 1277–1286.
Nagahara  N, Nishino  T. 1996. Role of amino acid residues in the ac-
tive site of rat liver mercaptopyruvate sulfurtransferase. CDNA cloning,
overexpression, and site-directed mutagenesis. The Journal of Biological
Chemistry 271, 27395–27401.
Nagahara  N, Okazaki  T, Nishino  T. 1995. Cytosolic mercaptopyruvate
sulfurtransferase is evolutionarily related to mitochondrial rhodanese.
Striking similarity in active site amino acid sequence and the increase in the
mercaptopyruvate sulfurtransferase activity of rhodanese by site-directed
mutagenesis. The Journal of Biological Chemistry 270, 16230–16235.
Nahar N, Rahman A, Ghosh S, Nawani N, Mandal A. 2017. Functional
studies of AtACR2 gene putatively involved in accumulation, reduction and/
or sequestration of arsenic species in plants. Biologia 72, 520–526.
Nakai  Y, Harada  A, Hashiguchi  Y, Nakai  M, Hayashi  H. 2012.
Arabidopsis molybdopterin biosynthesis protein Cnx5 collaborates with the
ubiquitin-like protein Urm11 in the thio-modification of tRNA. The Journal of
Biological Chemistry 287, 30874–30884.
Nakai Y, Nakai M, Hayashi H. 2008. Thio-modification of yeast cytosolic
tRNA requires a ubiquitin-related system that resembles bacterial sulfur
transfer systems. The Journal of Biological Chemistry 283, 27469–27476.
Nakamura  T, Yamaguchi  Y, Sano  H. 2000. Plant mercaptopyruvate
sulfurtransferases: molecular cloning, subcellular localization and enzymatic
activities. European Journal of Biochemistry 267, 5621–5630.
Nambi  S, Long  JE, Mishra  BB, Baker  R, Murphy  KC, Olive  AJ,
Nguyen  HP, Shaffer  SA, Sassetti  CM. 2015. The oxidative stress net-
work of Mycobacterium tuberculosis reveals coordination between radical
detoxification systems. Cell Host & Microbe 17, 829–837.
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
Narsai R, Law SR, Carrie C, Xu L, Whelan J. 2011. In-depth temporal
transcriptome profiling reveals a crucial developmental switch with roles for
RNA processing and organelle metabolism that are essential for germination
in Arabidopsis. Plant Physiology 157, 1342–1362.
Noma  A, Sakaguchi  Y, Suzuki  T. 2009. Mechanistic characterization
of the sulfur-relay system for eukaryotic 2-thiouridine biogenesis at tRNA
wobble positions. Nucleic Acids Research 37, 1335–1352.
Nomura H, Komori T, Kobori M, Nakahira Y, Shiina T. 2008. Evidence
for chloroplast control of external Ca2+-induced cytosolic Ca2+ transients
and stomatal closure. The Plant Journal 53, 988–998.
Pagani S, Forlani F, Carpen A, Bordo D, Colnaghi R. 2000. Mutagenic
analysis of Thr-232 in rhodanese from Azotobacter vinelandii highlighted the
differences of this prokaryotic enzyme from the known sulfurtransferases.
FEBS Letters 472, 307–311.
Pantoja-Uceda D, López-Méndez B, Koshiba S, et al. 2005. Solution
structure of the rhodanese homology domain At4g01050(175-295) from
Arabidopsis thaliana. Protein Science 14, 224–230.
Papenbrock  J, Guretzki  S, Henne  M. 2011. Latest news about the
sulfurtransferase protein family of higher plants. Amino Acids 41, 43–57.
Papenbrock J, Schmidt A. 2000. Characterization of two sulfurtransferase
isozymes from Arabidopsis thaliana. European Journal of Biochemistry 267,
5571–5579.
Paul  BD, Snyder  SH. 2012. H2S signalling through protein sulfhydration
and beyond. Nature Reviews. Molecular Cell Biology 13, 499–507.
Peer  WA, Baxter  IR, Richards  EL, Freeman  JL, Murphy  AS.
2006. Phytoremediation and hyperaccumulator plants. In: Tamas  MJ,
Martinoia  E, eds. Topics in current genetics. Molecular biology of metal
homeostasis and detoxification: from microbes to man. Berlin, Heidelberg:
Springer, 299–340.
Peltier JB, Ytterberg AJ, Sun Q, van Wijk KJ. 2004. New functions of
the thylakoid membrane proteome of Arabidopsis thaliana revealed by a
simple, fast, and versatile fractionation strategy. The Journal of Biological
Chemistry 279, 49367–49383.
Pérez-Pérez  ME, Mauriès  A, Maes  A, Tourasse  NJ, Hamon  M,
Lemaire  SD, Marchand  CH. 2017. The deep thioredoxome in
Chlamydomonas reinhardtii: new insights into redox regulation. Molecular
Plant 10, 1107–1125.
Phizicky EM, Hopper AK. 2010. tRNA biology charges to the front. Genes
& Development 24, 1832–1860.
Ploegman JH, Drent G, Kalk KH, Hol WG. 1978. Structure of bovine liver
rhodanese. I. Structure determination at 2.5 Å resolution and a comparison
of the conformation and sequence of its two domains. Journal of Molecular
Biology 123, 557–594.
Ranjan N, Rodnina MV. 2017. Thio-modification of tRNA at the Wobble
position as regulator of the kinetics of decoding and translocation on the
ribosome. Journal of the American Chemical Society 139, 5857–5864.
Sánchez-Bermejo E, Castrillo G, del Llano B, et al. 2014. Natural vari-
ation in arsenate tolerance identifies an arsenate reductase in Arabidopsis
thaliana. Nature Communications 5, 4617.
Schwarz  G, Mendel  RR. 2006. Molybdenum cofactor biosynthesis and
molybdenum enzymes. Annual Review of Plant Biology 57, 623–647.
Senkler  J, Senkler  M, Eubel  H, Hildebrandt  T, Lengwenus  C,
Schertl P, Schwarzländer M, Wagner S, Wittig I, Braun H-P. 2016. The
4154 | Selles et al.
mitochondrial complexome of Arabidopsis thaliana. The Plant Journal 89,
1079–1092.
Shen  J, Keithly  ME, Armstrong  RN, Higgins  KA, Edmonds  KA,
Giedroc  DP. 2015. Staphylococcus aureus CstB is a novel multidomain
persulfide dioxygenase-sulfurtransferase involved in hydrogen sulfide de-
toxification. Biochemistry 54, 4542–4554.
Shibuya N, Tanaka M, Yoshida M, Ogasawara Y, Togawa T, Ishii K,
Kimura H. 2009. 3-Mercaptopyruvate sulfurtransferase produces hydrogen
sulfide and bound sulfane sulfur in the brain. Antioxidants & Redox Signaling
11, 703–714.
Slade  WO, Werth  EG, McConnell  EW, Alvarez  S, Hicks  LM. 2015.
Quantifying reversible oxidation of protein thiols in photosynthetic or-
ganisms. Journal of the American Society for Mass Spectrometry 26,
631–640.
Sorrell DA, Chrimes D, Dickinson JR, Rogers HJ, Francis D. 2005.
The Arabidopsis CDC25 induces a short cell length when overexpressed
in fission yeast: evidence for cell cycle function. New Phytologist 165,
425–428.
Spadafora  ND, Doonan  JH, Herbert  RJ, Bitonti  MB, Wallace  E,
Rogers  HJ, Francis  D. 2011. Arabidopsis T-DNA insertional lines for
CDC25 are hypersensitive to hydroxyurea but not to zeocin or salt stress.
Annals of Botany 107, 1183–1192.
Su T, Xu J, Li Y, Lei L, Zhao L, Yang H, Feng J, Liu G, Ren D. 2011.
Glutathione-indole-3-acetonitrile is required for camalexin biosynthesis in
Arabidopsis thaliana. The Plant Cell 23, 364–380.
Takahashi H, Kopriva S, Giordano M, Saito K, Hell R. 2011. Sulfur as-
similation in photosynthetic organisms: molecular functions and regulations
of transporters and assimilatory enzymes. Annual Review of Plant Biology
62, 157–184.
Teschner  J, Lachmann  N, Schulze  J, Geisler  M, Selbach  K,
Santamaria-Araujo J, Balk J, Mendel RR, Bittner F. 2010. A novel role
for Arabidopsis mitochondrial ABC transporter ATM3 in molybdenum co-
factor biosynthesis. The Plant Cell 22, 468–480.
Tomizioli M, Lazar C, Brugière S, et al. 2014. Deciphering thylakoid sub-
compartments using a mass spectrometry-based approach. Molecular &
Cellular Proteomics 13, 2147–2167.
Vainonen JP, Sakuragi Y, Stael S, et al. 2008. Light regulation of CaS,
a novel phosphoprotein in the thylakoid membrane of Arabidopsis thaliana.
The FEBS Journal 275, 1767–1777.
Venne  AS, Solari  FA, Faden  F, Paretti  T, Dissmeyer  N, Zahedi  RP.
2015. An improved workflow for quantitative N-terminal charge-based frac-
tional diagonal chromatography (ChaFRADIC) to study proteolytic events in
Arabidopsis thaliana. Proteomics 15, 2458–2469.
Vojta L, Carić D, Cesar V, Antunović Dunić J, Lepeduš H, Kveder M,
Fulgosi  H. 2015. TROL-FNR interaction reveals alternative pathways of
electron partitioning in photosynthesis. Scientific Reports 5, 10085.
Weinl S, Held K, Schlücking K, Steinhorst L, Kuhlgert S, Hippler M,
Kudla  J. 2008. A plastid protein crucial for Ca2+-regulated stomatal re-
sponses. New Phytologist 179, 675–686.
Williams PN, Islam MR, Adomako EE, Raab A, Hossain SA, Zhu YG,
Feldmann J, Meharg AA. 2006. Increase in rice grain arsenic for regions
of Bangladesh irrigating paddies with elevated arsenic in groundwaters.
Environmental Science & Technology 40, 4903–4908.
Winter D, Vinegar B, Nahal H, Ammar R, Wilson GV, Provart NJ. 2007.
An “Electronic Fluorescent Pictograph” browser for exploring and analyzing
large-scale biological data sets. PLoS ONE 2, e718.
Downloaded from https://academic.oup.com/jxb/article/70/16/4139/5485845 by guest on 08 February 2023
Wollers S, Heidenreich T, Zarepour M, Zachmann D, Kraft C, Zhao Y,
Mendel RR, Bittner F. 2008. Binding of sulfurated molybdenum cofactor
to the C-terminal domain of ABA3 from Arabidopsis thaliana provides in-
sight into the mechanism of molybdenum cofactor sulfuration. The Journal
of Biological Chemistry 283, 9642–9650.
Xiong L, Ishitani M, Lee H, Zhu JK. 2001. The Arabidopsis LOS5/ABA3
locus encodes a molybdenum cofactor sulfurase and modulates cold
stress- and osmotic stress-responsive gene expression. The Plant Cell 13,
2063–2083.
Yadav PK, Yamada K, Chiku T, Koutmos M, Banerjee R. 2013. Structure
and kinetic analysis of H2S production by human mercaptopyruvate
sulfurtransferase. The Journal of Biological Chemistry 288, 20002–20013.
Yoshida K, Noguchi K, Motohashi K, Hisabori T. 2013. Systematic ex-
ploration of thioredoxin target proteins in plant mitochondria. Plant & Cell
Physiology 54, 875–892.
Zheng  L, White  RH, Cash  VL, Jack  RF, Dean  DR. 1993. Cysteine
desulfurase activity indicates a role for NIFS in metallocluster biosynthesis.
Proceedings of the National Academy of Sciences, USA 90, 2754–2758.
Zhong R, Thompson J, Ottesen E, Lamppa GK. 2010. A forward gen-
etic screen to explore chloroplast protein import in vivo identifies Moco
sulfurase, pivotal for ABA and IAA biosynthesis and purine turnover. The
Plant Journal 63, 44–59.